See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/258803594

Caligula, a Stardust Sulfide-Silicate Assemblage Viewed Through SEM,

NanoFTIR, and STXM

Article · March 2013

CITATIONS READS

2 54

9 authors, including:

Zack Gainsforth Alexander S McLeod

University of California, Berkeley Columbia University
115 PUBLICATIONS   1,597 CITATIONS    79 PUBLICATIONS   2,697 CITATIONS   

SEE PROFILE SEE PROFILE

Anna Butterworth Gerardo Dominguez

University of California, Berkeley California State University, San Marcos
132 PUBLICATIONS   2,881 CITATIONS    98 PUBLICATIONS   3,485 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Theory and interpretation of near-field optical microscopy View project

Isotopic Studies of Water Formation in Astrophysical Environments View project

All content following this page was uploaded by Zack Gainsforth on 19 August 2014.

The user has requested enhancement of the downloaded file.

44th Lunar and Planetary Science Conference (2013) 2332.pdf

CALIGULA, A STARDUST SULFIDE-SILICATE ASSEMBLAGE VIEWED THROUGH SEM, NANOFTIR AND STXM.
Zack Gainsforth1 , Alex McLeod2 , Anna Butterworth1 , Gerardo Dominguez3 , Dimitri Basov2 , Fritz Keilmann4 , Mark Thiemens2 ,
Tolek Tyliszczak5 , Andrew J. Westphal1 , 1 Space Sciences Laboratory, University of California at Berkeley, Berkeley CA 94720,
USA, 2 University of California at San Diego, La Jolla, CA 92093, USA, 3 California State University, San Marcos, San Marcos,
CA 92096-0001, USA, 4 LASNIX, Sonnenweg 32, 82152 Berg, Germany, 5 Advanced Light Source, Lawrence Berkeley National
Laboratory, Berkeley, CA 94720-8225, USA .

Introduction (named Caligula) is a 12 x 6 µm complex assemblage of

NASA’s Stardust mission returned samples from Comet
81/P Wild 2[1], comprising a broad diversity of materials in-
cluding amorphous condensation products and melt assem-
blages containing glasses[2]. Characterizing amorphous phases
is key to understanding the formation of these primitive solar
system constituents. Here we apply a new combination of tech-
niques to examine an amorphous phase in one such Stardust
sample, a grain named Caligula.
Use of NanoFTIR
Fourier transform infrared (FTIR) is powerful for study-
ing bond structures, but has limited spatial resolution. In
the case of the 10 µm silicate stretch, the diffraction limit
is theoretically about 5 µm, but practically close to 10 µm
in most cases. Particles returned by NASA’s Stardust mis-
sion have sub-micron internal structures, so FTIR is difficult
to apply. However, scanning near-field optical microscopy,
or nanoFTIR[3][4][5] is a relatively new atomic force mi-
croscopy, that can probe the infrared optical qualities of sam-
ples far below the usual diffraction limit. Just like FTIR, it
can potentially map amorphous silicate structures[8]. Spectra
are not equivalent to, but can theoretically be related in a one
to one fashion with traditional FTIR spectra. In our case we
probed the 10 µm silicate stretch with ≈ 100 nm spatial reso-
lution. Other promising applications include the examination
of carbonaceous and silicate materials in meteorites[6].
Experimental
The largest terminal particle of Stardust track C2035,5,105
pyrrhotite and heterogeneous silicate. It was embedded in
epoxy and the top ≈ 5 µm was ultramicrotomed into 80 nm
thick sections at the National Center for Electron Microscopy
(NCEM). The remaining bullet was analyzed by nanoFTIR
at the University of California in San Diego (UCSD) using a
Neaspec atomic force microscope with a broadband laser from
Toptica Photonics tuned to a 300 cm−1 bandwidth around the
10 µm silicate stretch (750-1100 cm−1 ). Difference-frequency
generation components were provided by Lasnix.
The bullet was then imaged in a Tescan Vega III scanning
electron microscope (SEM) at the University of California at
Berkeley (UCB) with an 80mm2 EDS detector from Oxford.
Quantitative EDS maps were taken at 10 keV and 200 pA which
gave a spatial resolution of around a few hundred nm, varying
by element. Figure 1 shows the SEM backscatter image and an
RGB image produced from Ca, Mg and Na SEM/EDS X-ray
maps.
The ultramicrotomed slices were imaged using scanning
transmission X-ray microscopy (STXM) at the Advanced Light
Source (ALS) in Berkeley, CA. Future work will involve TEM
analysis of the ultramicrotomed slices at NCEM.
Observations
Pyrrhotite: SEM/EDS shows that the pyrrhotite has com-
position Fe1−x S where x < 0.01 (2σ statistical) relative to
Canyon Diablo troilite. Some Si and O are present in the spec-
trum as well, but could possibly be due to the close proximity
of the silicate. Future transmission electron diffraction will

Figure 1: SEM images of Caligula. A) Backscattered electron
image taken at 10 keV shows the sulfide in the upper right,
and silicate in the lower left. B) Tricolor plot made from EDS
maps for Ca (red), Mg (green), and Na (blue). The Brehm-
strahllung in the Ca channel (red) shows the outline of the
sulfide. There is no Ca in the sulfide. Scalebars are 1 µm.
Figure 2: NanoFTIR images of Caligula. A) Whitebeam op-
tical image showing phases and phase boundaries in the in-
frared. B) Tricolor image constructed from three frames of a
spectrum image. Red is the 940 cm−1 frame, green is 993.33
cm−1 , and blue is 1043.33 cm−1 . The pixels are 100 nm in
width. Scalebars are 1 µm.
44th Lunar and Planetary Science Conference (2013)
determine the presence or absence of a pyrrhotite superlattice.
Non-stoichiometric silicate matrix: The silicate assem-
blage comprises a non-stoichiometric matrix with nanophase
inclusions. The matrix composition is (At %): O = 60.85, Na
= 6.52, Mg = 5.59, Al = 0.51, Si = 23.62, P = 0.63, S = 0.11,
K = 0.19, Ca = 0.22, Mn = 0.52, Fe = 1.22. The most variable
element is Na which varies down to 5.80 At %. An Mg K-edge
XANES spectrum acquired by STXM shows almost no struc-
ture, with a single 10 eV wide resonance enhancement that has
an optical density only ≈ 20% higher than the edge jump. No
EXAFS oscillations are visible in the subsequent 100 eV. Oth-
erwise, the spectrum looks close to a theoretical hydrogenic
model[7]. As described, the Mg XANES spectrum is strongly
suggestive of an amorphous coordination environment for the
Mg cation. NanoFTIR shows a peak at 950 cm−1 on the side
furthest from the sulfide (green region in Figure 2B), while a
peak at 1040 cm−1 is present everywhere but lower in intensity
away from the sulfide. In both cases, the FTIR peaks are broad
and poorly resolved, which is consistent with the hypothesis
that the phase is amorphous.
Pyroxene: The largest silicate inclusion is ≈ 1 µm across
(green in Figure 1B, red in Figure 2B). It is euhedral with
(At %): O = 60.41, Na = 1.75, Mg = 15.05, Si = 21.68,
Ca=0.67, Fe=0.44. The composition matches enstatite with
minor Ca and Na (≈ En93 Wo3 Ae4 ). Mg K-edge XANES
going 100 eV beyond the edge gives a spectrum with three
distinct resonance enhancements. The central peak is at 1314
±1 eV, with relative peaks at ∆E = −4.2 and +4.8 eV, and
two EXAFS oscillations, which compares favorably against an
enstatite standard. The nanoFTIR spectrum shows two strong,
well defined peaks at 950 cm−1 and 1060 cm−1 and a weaker
peak at 1010 cm−1 .
NanoFTIR Acquisition
Figure 2A shows a white beam image of Caligula produced
without acquiring a spectrum, but instead using the full band-
width of the laser (in this case about 300 cm−1 ). This method
allows fast acquisitions with excellent contrast. However, the
resulting image is not quantitative except in determining phase
locations and boundaries.
Figure 2B is an RGB image produced from three energies
in a spectral stack containing 68 energies spanning 850-1073
cm−1 . The entire stack required about 6 hours to complete and
was acquired during a period of exceptional laser stability. The
image is skewed slightly due to drift in the nano positioning
stage over the 6 hour acquisition, but corresponds to the bottom
half of Figure 2A.
The three energies were chosen to maximize contrast be-
tween the phases involved. The 1040 cm−1 peak (blue) is
present in the non-stoichiometric silicate matrix but is off res-
onance for the pyroxene inclusion, and mostly tracks the ma-
trix. The 940 cm−1 peak (red) is present in both the matrix
and the pyroxene inclusion, but is very much more intense in
the inclusions, hence they are red. The 993 cm−1 peak (green)
resides in a minima between the 940 and 1043 cm−1 peaks,
and therefore provides good contrast for variations in the ma-
trix without being swamped by the 940 cm−1 peak found in
the inclusions. The pyrrhotite is reflective in IR and shows
white.
View publication stats
2332.pdf
In Figure 2B, we can clearly see a gradient in the matrix
showing the gradual variation in the silicate character from
one side of the grain to the other. It is seen as a variation
from blue to green as the 1040 cm−1 (blue) peak intensity
varies. Comparison against the EDS map in Figure 1B shows
that the silicate is varying in the region containing sub-micron
inclusions as well as a region without inclusions. Therefore, it
appears to be tracking a property of the matrix itself.
Integrating methods
The combination of SEM/TEM, STXM, and nanoFTIR
are necessary to understand amorphous silicates in Stardust.
SEM/TEM provides characterization of the material and iden-
tification of related crystallites.
STXM provides ordering information of many cations and
the structural elements silicon, oxygen and aluminum. Here,
we saw in the Mg XANES that there was no clear order around
the Mg cations in the non-stoichiometric silicate matrix.
Like FTIR, nanoFTIR appears to provide a very sensitive
tracer of silicate structure making the ordering process visible
before cation ordering becomes visible in XANES spectra.
In the matrix, the silicate stretch contained two peaks whose
positions did not change, but whose intensities did. While
we cannot interpret this quantitatively yet, we know at least
that there is a variability from one side of the particle to the
other, even though the Mg ordering has not occurred. Thus
conjunction of STXM and nanoFTIR data provides the relative
ordering sequence of elements in the glass.
Future work
Caligula appears to be a complex assemblage of amor-
phous silicate with pyroxene inclusions and possibly other
silicate inclusions in close association with a large sulfide and
likely fine grained sulfides as well. Further TEM analysis
should elucidate the identity and crystallinity of the phases
and allow more quantitative analysis of both the XANES and
nanoFTIR results. Once this is accomplished, we hope to form
a plausible formation mechanism for Caligula in the context
of its cometary history and/or condensation processes, etc.
Acknowledgments
UCB/SSL funded by a NASA Stardust Participating Sci-
entist grant. The UCSD effort was supported by NASA LARS
grant NNX11AF24G. The ALS and NCEM are is supported by
the Director, Office of Energy Research, Office of Basic Energy
Sciences, Materials Sciences Division of the U.S. Department
of Energy, under Contract No. DOE-AC03-76SF00098.
References
[1] Brownlee, D. E., et. al., (2006) Science, 314, 1711.
[2] Joswiak, D., et. al., (2012) Meteoritics & Planet. Sci., 47,
471. [3] Keilmann, F., Hillenbrand, R., (2004) Phil. Trans. R.
Soc. A, 362, 787-805. [4] Keilmann, F., Amarie, S., (2012)
J. Infrared Milli. Terahz Waves, 33, 479-484. [5] Amarie, S.,
Keilmann, F., (2011) Phys. Rev. B, 83, 45404. [6] McLeod,
A. S., et al, (2013) 44th LPSC, submitted. [7] Egerton, R.
F., (1996) 2nd ed., Plenum Press. [8] Mutschke, H., (1998)
Astron. Astrophys., 333, 188-198